The inability to exert appropriate inhibitory control over one's behavior is an important symptom of many psychiatric illnesses. The stop signal paradigm, which involves withholding a prepotent motor response to a signal, has proved useful in assessing this deficit in children with a range of disorders, particularly attention deficit hyperactivity disorder (ADHD). In order to adapt the paradigm for use in fMRI studies, it is important to address two methodological issues: 1) the substitution of a visual stop signal for the more commonly-employed auditory stop signal, and 2) the relative merits of the traditional stop task (subjects withhold a motor response to the "stop" signal) vs. the stop-change task (subjects execute alternative motor responses to the "change" and "go" signals). The first aim of this protocol is to conduct a pilot fMRI study of the stop and stop-change tasks in control adults and adolescents. Secondly, behavioral testing is being conducted in children with bipolar disorder, in order to facilitate study of their deficits in impulse control. The pilot fMRI study of control adults indicates that anterior cingulate (AC) activation occurs in all stop and change trials, whether they are successful or not. This result is more consistent with theories implicating the AC in resolution of conflict regarding alternative response choices, rather than in error monitoring. Orbitofrontal cortical (OFC) activation occurs in both change and stop trials, contradicting theories ascribing OFC activation to inhibition alone. With regard to children with BPD, behavioral data indicate that such children's performance on the stop task does not differ from that of controls, but that their performance on the change task is deficient. That is, children with BPD take considerably longer to accomplish the response reversal required to complete the task; comorbid attention deficit hyperactivity disorder does not appear to account for this deficit. These data are congruent with other data that we have collected indicating response reversal difficulties in children with BPD and indicates possible dysfunction in neural circuits involving the AC and/or OFC. fMRI data on control adolescents has been collected and is being analyzed currently, and fMRI data on children with BPD is being collected.